MCC Lec01

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

3/1/23

Introduction to Cloud Computing


Dr Nazanin Mohammadi

References:
Harkut, D. G. (2018). Introductory Chapter: Cloud Computing. Cloud Computing-Technology and Practices, IntechOpen.

Comer, D. E. (2021). The Cloud Computing Book: The Future of Computing Explained, Chapman and Hall/CRC.

• Definition from NITS (National Institute of Standards and


Technology)

“Cloud computing is a model for enabling ubiquitous,


What is Cloud convenient, on-demand network access to a shared pool
of configurable computing resources (e.g., networks,
Computing servers, storage, applications, and services) that can be
rapidly provisioned and released with minimal
management effort or service provider interaction. This
cloud model is composed of five essential characteristics,
three service models, and four deployment models.”

1
3/1/23

On-
Sophisticated
Demand Technologies

Device-
Independent

Scalable Reliable
Services

Evolution of Computing Models

Utility Computer Grid Cloud


Computing Cluster Computing Computing

2
3/1/23

Utility Computing
• Service provisioning computing model

• Based on a conventional utilities pay-per-use model, such as


electricity/gas/water

• Very similar to cloud computing but different in:


• The geographical location of the service and
infrastructure
• Cost
• Resource demand- the cloud is better for more resource
demand application

https://www.spiceworks.com /tech/cloud/articles/what-is-utility-com puting/

Computer Cluster
• A set of same linked (in
LAN)computers that appear as one
powerful computer.
• Failure of one computer won’t
affect the performance of the
computer cluster
• Mostly used for HPC (High-
Performance Computing) and HA
(high availability) applications
• Usually used for a specific function:
• Load balancing
• High availability
• High performance
• Large-scale processing
• Runs by a central management
software

https://www.suse.com /suse-defines/definition/com puter-cluster/

3
3/1/23

Grid Computing

• A set of unlike computers


linked over different
geographical locations
• Distributed resource
management
• Used mostly for tasks such as:
• Predictive modelling
• Automation
• Simulations
• Distributed resource
management

Cloud Computing
• Sharing computing resources
• Using software and hardware to
share resources over the network
• The word Cloud is used to
describe the network and mainly
the Internet
• Anywhere, any time, on any
device (with an Internet
connection)
• Web-based applications

4
3/1/23

Why Cloud Computing


• Motivation
• Technological
• Economical
• More data
• More processing power
• Analysis and Prediction
• This will cost more for in-house computing

The Structure of Cloud Computing

NETWORK DATA CENTRE VIRTUAL


ENVIRONMENT

10

5
3/1/23

Data Centre
• Buildings • Racks • PoD (point of
• Like a warehouse • Holding equipment Delivery)
• Open plan, no wall • Placed side by side • Servers
• Example : • Networking
• Flexential’s ComPark data • Storage
centre in Denver
• PDUs
• Single room 13800
Square meter • Management
• 216 meter facilities
• Media companies’ data
centres can be > 93000
Square meters

11

Networking
• Interconnected devices
• Different protocol to communicate between devices and the Internet

12

6
3/1/23

• Reason for having elasticity in cloud computing


• Technology is older than the cloud
• Software emulation
• Replicates the behaviour of another computer to minimize the
need of developing a compiler for each type of computer.
Example: WINE runs Windows app on Mac or Linux.

Virtual • It adds overhead


• Para-Virtualization

environment
• An early version of virtualization
• Uses hypervisor to run multiple OS
• Requires altering the OS code
• Full-Virtualization
• Same as Para-virtualization
• Does not required OS code changes and is faster

13

Private cloud
Types of cloud
Public Cloud

Hybrid cloud

14

7
3/1/23

Private Cloud

• Also known as Internal Cloud


• Utilized and operated only by one organization
• Internal Data Centre
• Spreads the computing process to all servers to avoid underutilization
and overutilization
• Control over underlying infrastructure
• Low Latency (on-premises facility)
• More flexibility as there is no lock-in contract

15

Public Cloud
• Full Cloud Computing
• The organization leases services including infrastructure
• Economic option
• Providers operate on a large scale therefore they can use technologies
to reduce the cost such as SDN (Software Defined Network)
• Providers have large-scale expertise and can share them with clients
• Can offer AI and ML
• Migration services

16

8
3/1/23

Hybrid Cloud
• Customized cloud system
• Combination of public and private cloud
• Control for an organization with regulations and assign some tasks as
private cloud
• In case of high computing load, the public section can be used to
overcome insufficient resource problem exits in private cloud
• Security management is challenging

17

Cloud Services

• Software as a Service (SaaS)

• Platform as a Service (PaaS)

• Infrastructure as a Service (IaaS)

18

9
3/1/23

SaaS
• Provides software with monthly fee payment
• The application runs on a cloud
• Data is stored on the cloud
• e.g., Microsoft’s Office 365

• Characteristics:
• Universal access: anytime/any device
• Guaranteed synchronization: any changes in a file are applied and saved.
Therefore, data appears the same if the user utilizes multiple devices to login
• No delay caused by a power outage

19

PaaS
• Cloud subscriber uses this facility to develop, build, and deploy
software on the cloud
• Compiler
• Middleware
• Program libraries
• Runtime system (e.g., Java, .NET)
• …

20

10
3/1/23

IaaS
• Provides infrastructure such as:
• Buildings
• Power
• Aircon
• Servers
• Networking
• Storage
• Load balancer
• Data backup
• Network security
• …

21

Advantages
• Reliability
• Data recovery
• Manageability
• Data centralization
• Cost saving
• Device independence
• Strategic Edge

22

11
3/1/23

Disadvantages
• Uptime

• Low bandwidth

• Security

• Host lock-in

• Limited flexibility

• Minimal Control

• Incompatibility

23

Main Challenges
• Availability
• High-speed Internet is required

• Security
• Security implementation at all levels

24

12
3/1/23

25

13

You might also like